# SerpKite

> SerpKite provides search and web content APIs for AI agents: clean JSON or Markdown, official TypeScript/Python/Go SDKs, native LangChain, CrewAI, Vercel AI SDK and MCP, credits that never expire.

- Base URL: https://api.serpkite.com. Every route is under `/v1` (`/v1/search`, `/v1/news`, `/v1/images`, `/v1/videos`, `/v1/maps`, `/v1/places`, `/v1/place`, `/v1/reviews`, `/v1/shopping`, `/v1/scholar`, `/v1/patents`, `/v1/autocomplete`, `/v1/webpage`, `/v1/answer`, `/v1/extract`, `/v1/map`, `/v1/crawl`, `/v1/research`, `/v1/chatgpt`, `/v1/gemini`, `/v1/copilot`, `/v1/ai_mode`, `/v1/brave_ai`, `/v1/monitors`, `/v1/rank`, `/v1/account`, `/v1/status`). `GET /customsearch/v1` serves apps moving off the Google Custom Search JSON API.
- Auth: `Authorization: Bearer skt_live_…` (GET may use `?api_key=`). Search verticals and webpage accept POST JSON or GET query params. Site/job creation uses POST; task polling uses GET, monitor updates PATCH, cancellation/deletion DELETE. Search parameters: `q`, `country`, `language`, `location`, `num`, `page`, `time`, `device`, `format` (json|compact|markdown), `fields` (e.g. `results.title,results.link,knowledge_graph`), `include_content`, `highlights`, `include_answer`, `include_domains`, `exclude_domains`, `boost_domains`, `start_date`, `end_date`, `max_age`. Unknown parameters return 400 naming the replacement.
- Response envelope for every vertical: `request`, `results` (the primary list), vertical extras (`answer_box`, `ai_overview`, `knowledge_graph`, `people_also_ask`, `related_searches`, `top_stories`, `places`…) and `meta` (`request_id`, `credits_used`, `cached`, `latency_ms`). All keys snake_case. Errors: `{"error":{"code","message","request_id"}}`.
- Places: `/v1/maps` and `/v1/places` return Google Maps search results (address, coordinates, categories, rating, hours, phone, website, `place_id`/`cid`/`fid`) in the request `language`; pass `location` or `ll` (`@lat,lng,zoomz`) for local queries. `/v1/place` takes one `place_id`, `cid` or `fid` and returns that place's full listing with `country`, `timezone`, structured `weekly_hours` (local `{open, close}` per weekday), `international_phone_number`, `description` and `photos` (1 credit; empty and not billed when no place matches). `engine: "yandex"` on maps/places answers from Yandex Maps (Russia and the CIS; places carry `oid` and `yandex_maps_url`), and `/v1/place` takes an `oid` instead of the Google ids.
- Providers: Google by default; explicit `engine` overrides the key fallback setting. `auto` permits enabled fallbacks, a provider list restricts them, and search-only `consensus` merges independent indexes with `results[].sources`. Keep `meta.engine` for provenance. Domain filters and boosts cannot be combined with consensus.
- Site content: `POST /v1/extract` reads 1–20 HTML/PDF URLs (1 credit per successful live URL, 0.5 cached; check `failed`). `POST /v1/map` discovers site URLs (1 credit, free if empty). `POST /v1/crawl` reserves `limit` credits and returns a task; poll/cancel `/v1/crawl/{id}` or receive `crawl.completed`. Crawl results expire after 24 hours. Site formats select content inside JSON.
- AI (written by Anthropic Claude Haiku 5.5; typed checks by OpenAI's Decisions API; a failed AI step is never billed): `include_answer: true` on /v1/search and /v1/news adds `answer` {`text` with [n] markers, `citations`} (+2 credits, +4 with include_content). `POST /v1/answer` {`q`, `depth`: fast|standard, `num_sources` 1–5, `stream` (SSE: results, delta, done|error), `output_schema`} returns `answer`, `citations`, `results`; the search part is billed like a search and the model adds 2 (fast) or 4 (standard); `503 ai_unavailable` refunds the model part. /v1/extract `summary: true` (+4/page) and `output_schema` + `prompt` (+4/page, JSON in `results[].output`). `POST /v1/research` {`input`, `effort`: lite 30 | standard 75 | deep 200 credits flat, `output_schema`, `webhook_url`} returns a task; poll/cancel `/v1/research/{id}` or receive `research.completed`; `result.report` is Markdown with [n] citations. Query and page text go to Anthropic and OpenAI under API terms that exclude training.
- AI assistants (asked as a logged-out visitor in a standard browser; never signed in): `POST /v1/chatgpt` (ChatGPT, 4 credits, 10–30 s); `POST /v1/gemini` (Google Gemini, 3 credits, 5–30 s); `POST /v1/copilot` (Microsoft Copilot Search, 3 credits, 15–40 s); `POST /v1/ai_mode` (Google AI Mode, 2 credits, 5–45 s); `POST /v1/brave_ai` (Brave Search AI answers (Ask Brave), 2 credits, 1–10 s). Body: `prompt` (≤4000 chars), `country`, `language` (not chatgpt), `web_search` (chatgpt, default true), `track` (≤20 brands or domains), `include_html`, `format`, `fields`, `max_age` (answers cached ≤1 h, half price). Returns `text`, `markdown` ([n](link) citations matching `sources[].position`), `model`, `sources` (`cited` flag), `search_queries`, `products`, `places`, `related_questions`, and `mentions` {`term`, `mentioned`, `cited`, `sources`, `products`, `text_position`} with `track`. Failed, refused and empty answers are free; batches charge the full price. /v1/search returns Google's AI Overview as `ai_overview` when shown. Perplexity, Grok and Meta AI require sign-in and are not offered.
- Monitors: `POST /v1/monitors` saves search/news queries, a webpage URL, or an AI assistant prompt (`endpoint`: chatgpt|gemini|copilot|ai_mode|brave_ai, with `prompt` and `track`; every run reports the answer with `new_sources` and `mentions`). Read `/v1/monitors/{id}/runs` or receive `monitor.results`; each run is billed like its request, even when nothing new is found. Run result bodies expire after 24 hours. Resume paused monitors with PATCH `{"active":true}`; run-now requires active. Search monitors report new URLs, not rank changes.
- Batches: `POST /v1/batches` with `{"endpoint":"search","requests":[…up to 100],"webhook_url":"…"}` queues half-price jobs (AI assistant answers keep their full price); poll `GET /v1/batches/{id}` or receive a signed `batch.completed` webhook.
- 1 credit = 1 Google results page (10 results). `num=100` = 7 credits. Autocomplete costs 0.5; batch jobs and search cache charges 0.5×. `include_content` fetches still cost 1 credit per successful page on a search cache hit; highlights add no charge. Failed, empty and blocked requests are free.
- Packs from $10 for 10,000 credits down to $0.60 per 1,000 (largest pack $300); larger volumes on request (https://serpkite.com/enterprise). Free: 1,000 credits on signup, +1,500 for linking GitHub or Google, 1,000 every month.
- SDKs (key from SERPKITE_API_KEY): TypeScript `npm install serpkite`, Python `pip install serpkite` (sync + async), Go `go get github.com/serpkite/serpkite-go`, LangChain `pip install langchain-serpkite`, CrewAI `pip install "serpkite[crewai]"`, Vercel AI SDK tools `npm install @serpkite/ai-sdk`. Source: https://github.com/serpkite/serpkite-js, https://github.com/serpkite/serpkite-python, https://github.com/serpkite/serpkite-go, https://github.com/serpkite/langchain-serpkite, https://github.com/serpkite/serpkite-ai-sdk.
- Remote MCP server: https://api.serpkite.com/v1/mcp (streamable HTTP, OAuth or Bearer auth, 23 tools). OpenAPI: https://serpkite.com/openapi/serp-api.yaml. All docs in one file: https://serpkite.com/llms-full.txt.

## Docs: Getting started

- [SerpKite documentation](https://serpkite.com/docs.md): Search APIs, cited AI answers, page and PDF extraction, crawls, deep research and monitors for AI agents. Quickstart, endpoint reference, guides and MCP.
- [Quickstart](https://serpkite.com/docs/quickstart.md): Create an API key, install an SDK or use curl, run your first Google search with SerpKite, read the response and check your credit balance. About five minutes.
- [Authentication](https://serpkite.com/docs/authentication.md): Every SerpKite API call is authenticated with a secret API key, sent as a Bearer token in the Authorization header, or on GET requests as a query parameter.
- [Credits and billing](https://serpkite.com/docs/credits-and-billing.md): SerpKite is prepaid. You buy credit packs that never expire, each call spends a known number of credits, and failed or empty calls are refunded automatically.
- [Rate limits](https://serpkite.com/docs/rate-limits.md): SerpKite limits requests per second per API key. Free keys get 5 per second; a pack raises every key to that pack's rate. Over the limit: 429 with Retry-After.
- [Errors](https://serpkite.com/docs/errors.md): Every SerpKite error uses one JSON body with a stable code, a readable message and a request ID. Errors are never billed. Every code and what to do about it.
- [Response headers](https://serpkite.com/docs/response-headers.md): Every billed SerpKite response tells you its cost, your remaining balance, whether it came from cache, how long it took and about how many LLM tokens it holds.

## Docs: Core concepts

- [Common parameters](https://serpkite.com/docs/parameters.md): Every request parameter the SerpKite search endpoints accept, with types, defaults and allowed values, grouped by what they control.
- [Output formats](https://serpkite.com/docs/output-formats.md): Get Google results as full JSON, compact JSON or LLM-ready Markdown, and trim them with fields projection. The X-Tokens-Estimate header tells you the size.
- [Localization](https://serpkite.com/docs/localization.md): Search Google as a user in a specific country, language and city with country, language, location, uule and ll. Code tables for common countries and languages.
- [Pagination and depth](https://serpkite.com/docs/pagination-and-depth.md): Page through Google results with page, or fetch up to the top 100 in one call with num. num=100 is a depth bundle that costs 7 credits instead of 10.
- [Caching](https://serpkite.com/docs/caching.md): Accept a recent cached result with max_age and pay half the credits on a hit. How freshness, cache keys, X-Cache and meta.cached work.
- [Search providers](https://serpkite.com/docs/providers.md): When Google is unavailable, engine=auto or a per-key setting lets SerpKite fall back to Brave, Bing, Yahoo, DuckDuckGo, Mojeek or Yandex, named in meta.engine.
- [Batch requests](https://serpkite.com/docs/batch.md): Queue up to 100 requests per call with POST /v1/batches at half the credits. Poll GET /v1/batches/{id}, wait with the SDKs, or receive a signed webhook.
- [Webhooks](https://serpkite.com/docs/webhooks.md): Get finished batch jobs, crawls and new monitor results as signed POSTs: payload, headers, HMAC-SHA256 checks in Node.js, Python, Go and PHP, and retries.
- [Page content](https://serpkite.com/docs/include-content.md): Get the top results as Markdown in the same /v1/search call with include_content: one request instead of a search plus N page fetches, +1 credit per page.
- [Search filters](https://serpkite.com/docs/search-filters.md): Restrict, exclude or boost domains, bound results by publish date, read typed published_at dates and get query-ranked highlights instead of whole pages, free.
- [AI assistants](https://serpkite.com/docs/ai-assistants.md): Ask ChatGPT, Gemini, Copilot Search, Google AI Mode and Brave's AI answer by API as a logged-out user: answers, cited sources, products and brand mentions.

## Docs: Endpoints

- [Google Search](https://serpkite.com/docs/endpoints/search.md): Organic results, knowledge graph, answer box, AI Overview, People Also Ask, related searches, top stories and sitelinks.
- [Google News](https://serpkite.com/docs/endpoints/news.md): Google News results for a query: headline, source, publish date, snippet and thumbnail. Combine with time to watch a topic over the last hour, day or week.
- [Google Images](https://serpkite.com/docs/endpoints/images.md): Google Images results as JSON: the original image URL, width and height, thumbnail, title, and the page and domain it was found on, by country and language.
- [Google Videos](https://serpkite.com/docs/endpoints/videos.md): Google Videos results as JSON: title, link, channel, duration, publish date, snippet and thumbnail for each video, filterable by time, domain or date range.
- [Google Maps](https://serpkite.com/docs/endpoints/maps.md): Google Maps search: places with address, coordinates, categories, rating, hours, phone, website and place_id, in any language.
- [Google Places](https://serpkite.com/docs/endpoints/places.md): Local results for a search: the same Google Maps places and fields as /v1/maps, including address, coordinates, rating, category, hours, phone and place IDs.
- [Google Place Details](https://serpkite.com/docs/endpoints/place.md): One place by place_id, cid or fid (or a Yandex Maps oid): address, country, time zone, structured weekly hours, phones, website, description and photos.
- [Google Reviews](https://serpkite.com/docs/endpoints/reviews.md): The first 10 reviews of a place (by place_id, cid or fid), most relevant first, 10 per credit. Page with next_page_token.
- [Google Shopping](https://serpkite.com/docs/endpoints/shopping.md): Google Shopping products as JSON: the price as shown and as a number, currency, merchant, rating, review count, delivery info and product ID.
- [Google Scholar](https://serpkite.com/docs/endpoints/scholar.md): Google Scholar papers as JSON: title, authors and publication, year, citation count, snippet and a direct PDF link when one exists, by country and language.
- [Google Patents](https://serpkite.com/docs/endpoints/patents.md): Patents with publication number, title, assignee, inventor, priority, filing, grant and publication dates, and a PDF link.
- [Google Autocomplete](https://serpkite.com/docs/endpoints/autocomplete.md): Google's query suggestions for a prefix as a JSON list, by country and language. Useful for keyword research and for expanding a query before a search.
- [Webpage to Markdown](https://serpkite.com/docs/endpoints/webpage.md): Fetch any public URL (HTML or PDF) as clean Markdown and text with its title, description and metadata. Pages that need JavaScript are rendered in a browser.
- [Custom Search (CSE-compatible)](https://serpkite.com/docs/endpoints/customsearch.md): For apps moving off the Google Custom Search JSON API: same query params (key, cx, q, start, num…), same items[] and searchInformation shape.
- [Rank](https://serpkite.com/docs/endpoints/rank.md): Where does a domain rank for a keyword? Returns the best organic position in the top N and every matching result.
- [Answer](https://serpkite.com/docs/endpoints/answer.md): A cited answer to a question, written by Claude from live Google results and the top pages, with [n] citations. Streams over SSE or fills your JSON schema.
- [Extract](https://serpkite.com/docs/endpoints/extract.md): Read up to 20 URLs (HTML or PDF) per call as Markdown, with ranked highlights, an AI summary or JSON for your schema. Failed URLs are listed and cost nothing.
- [Map](https://serpkite.com/docs/endpoints/map.md): Lists a site's URLs from robots.txt, sitemaps and the start page's links, cleaned and de-duplicated, filtered by path regexes and optionally ranked by a phrase.
- [Crawl](https://serpkite.com/docs/endpoints/crawl.md): Async crawl of one site: up to 1,000 pages and 10 link hops from links and sitemaps, robots.txt respected. Poll GET /v1/crawl/{id} or get a signed webhook.
- [Research](https://serpkite.com/docs/endpoints/research.md): Async deep research: Claude plans Google searches, reads and ranks the pages, fills gaps and writes a Markdown report with numbered citations to its sources.
- [ChatGPT](https://serpkite.com/docs/endpoints/chatgpt.md): Ask ChatGPT as a logged-out user and get its answer, the sources it cited, product cards, local places and mentions of the brands and domains you track.
- [Gemini](https://serpkite.com/docs/endpoints/gemini.md): Ask Google Gemini as a logged-out user and get its answer as text and Markdown, with the sources it grounded on and the model it reports.
- [Copilot Search](https://serpkite.com/docs/endpoints/copilot.md): Ask Microsoft Copilot Search on Bing and get its answer with the sources it cited and the sponsored product cards beside it, for any Bing market.
- [Google AI Mode](https://serpkite.com/docs/endpoints/ai_mode.md): Ask Google AI Mode, Google's conversational search, as a logged-out user and get its answer with the sources it cited, for the country you choose.
- [Brave AI](https://serpkite.com/docs/endpoints/brave_ai.md): Ask Brave Search's AI answer and get it with cited sources, the searches it ran to write it and the products it showed, in 1 to 10 seconds.
- [Monitors](https://serpkite.com/docs/endpoints/monitors.md): Saved search or news requests that run hourly, daily or weekly and report only new results, or webpage URLs reported when they change, with signed webhooks.
- [Batches](https://serpkite.com/docs/endpoints/batches.md): Queue up to 100 requests for one endpoint at half price, then poll GET /v1/batches/{id} or get a signed webhook.
- [Account](https://serpkite.com/docs/endpoints/account.md): A free call returning the credit balance, rate limit, plan and this month's usage of the account behind the calling key, plus the key's limit when it has one.
- [Status](https://serpkite.com/docs/endpoints/status.md): Public live status: request volume, success rate and p50/p95 latency per endpoint over the last hour.

## Docs: Guides

- [Google Search API in Python](https://serpkite.com/docs/guides/google-search-python.md): Get Google search results in Python with the serpkite package: first query, country and city, paging, asyncio, CSV export, Markdown for LLMs and error handling.
- [Web search for agents with tool calling](https://serpkite.com/docs/guides/agents-tool-calling.md): Define a SerpKite search tool, retrieve query-ranked passages, and add page extraction or site discovery. Python and TypeScript code with schemas and budgets.
- [Build a RAG pipeline on live search](https://serpkite.com/docs/guides/rag-pipeline.md): Retrieve live search results and query-ranked passages, extract HTML pages or PDFs, and build an answer prompt with source URLs. Python and TypeScript examples.
- [Get cited answers from Google results](https://serpkite.com/docs/guides/cited-answers.md): Add a cited answer written by Claude to any Google search, or call /v1/answer to read the top pages first, stream it over SSE or get JSON for your schema.
- [Extract structured JSON from web pages](https://serpkite.com/docs/guides/structured-extraction.md): Turn web pages and PDFs into JSON with /v1/extract: send a JSON Schema and a prompt, get each page's data back. Classification, summaries, pricing and failures.
- [Run deep research with cited reports](https://serpkite.com/docs/guides/deep-research.md): Async deep research with /v1/research: Claude plans Google searches, reads and ranks the pages and writes a cited report. Efforts, polling, webhooks, pricing.
- [Ingest a website or a set of URLs](https://serpkite.com/docs/guides/site-ingestion.md): Discover URLs with map, read HTML pages and PDFs with extract, or crawl a site asynchronously. SDK examples, partial failures, polling and credit reservations.
- [Monitor news, search results and page changes](https://serpkite.com/docs/guides/scheduled-monitoring.md): Schedule search, news, webpage and AI assistant checks with monitors: run history, signed notifications, baselines, pause and resume, and budgets.
- [Track your brand in AI answers](https://serpkite.com/docs/guides/ai-visibility-tracking.md): Measure how ChatGPT, Gemini, Copilot, AI Mode and Brave answer about your brand: prompt sets, track and mentions, batches, monitors and run-to-run variance.
- [Choose search providers for your workflow](https://serpkite.com/docs/guides/provider-selection.md): Choose Google-only search, automatic fallback, a provider list or consensus. Keep rank measurements consistent and see which provider answered each result.
- [Rank tracking](https://serpkite.com/docs/guides/rank-tracking.md): Track a domain's Google top 100 rankings for many keywords by country, city and device with POST /v1/rank, num=100 (7 credits), half-price batches and webhooks.
- [Migrate from Google Custom Search](https://serpkite.com/docs/guides/migrate-from-google-cse.md): Google shuts down the Custom Search JSON API on January 1, 2027. SerpKite's GET /customsearch/v1 takes the same parameters and returns the same response shape.

## Docs: Tools and SDKs

- [MCP server](https://serpkite.com/docs/mcp.md): Give Claude, Cursor, VS Code, ChatGPT or any MCP client live Google search through SerpKite's remote MCP server: one URL, OAuth sign-in or an API key, 23 tools.
- [Official SDKs](https://serpkite.com/docs/sdks.md): Official SerpKite SDKs for TypeScript, Python (sync and async) and Go, plus LangChain, CrewAI and Vercel AI SDK packages: install, options, errors and batches.
- [OpenAPI spec](https://serpkite.com/docs/openapi.md): SerpKite's API is described by an OpenAPI 3.1 document: generate typed clients, import it into Postman, Insomnia or Bruno, or give it to an LLM.
- [API changelog](https://serpkite.com/docs/changelog.md): Changes to the SerpKite API contract, newest first. Additive changes ship without notice; anything breaking is announced here in advance.

## Docs: Account and dashboard

- [API keys](https://serpkite.com/docs/api-keys.md): Create as many SerpKite API keys as you need, give each a monthly credit limit, and rotate or revoke them without downtime. Keys are stored only as hashes.
- [Spend controls](https://serpkite.com/docs/spend-controls.md): Cap what your account spends per month, get emailed before the balance runs out or the cap is near, and get a one-click checkout link when it's time to top up.
- [Team](https://serpkite.com/docs/team.md): Invite teammates to manage keys, watch usage and buy credits on one shared SerpKite balance, while the account owner keeps control of settings and the account.
- [Privacy and data retention](https://serpkite.com/docs/privacy-and-data-retention.md): SerpKite never logs query text; failed requests keep theirs for a day to debug. Results are cached briefly under a hashed key; usage records are metadata only.

## SDKs

- [TypeScript: serpkite on npm](https://github.com/serpkite/serpkite-js): `npm install serpkite`, zero dependencies, Node 18+, Bun, Deno and edge.
- [Python: serpkite on PyPI](https://github.com/serpkite/serpkite-python): `pip install serpkite`, `SerpKite` and `AsyncSerpKite`.
- [Go module](https://github.com/serpkite/serpkite-go): `go get github.com/serpkite/serpkite-go`.
- [LangChain: langchain-serpkite](https://github.com/serpkite/langchain-serpkite): search tools, retriever and webpage loader.
- [CrewAI: serpkite[crewai]](https://github.com/serpkite/serpkite-python): `SerpKiteSearchTool`.
- [Vercel AI SDK: @serpkite/ai-sdk](https://github.com/serpkite/serpkite-ai-sdk): `npm install @serpkite/ai-sdk`, `serpkiteTools()` gives generateText, streamText and agents 21 tools (search, news, maps, place, webpage, extract, crawl, answer, research…). AI SDK 5–7, Next.js and edge.

## APIs

- [Google Search API](https://serpkite.com/apis/google-search-api)
- [Google News API](https://serpkite.com/apis/google-news-api)
- [Google Images API](https://serpkite.com/apis/google-images-api)
- [Google Videos API](https://serpkite.com/apis/google-videos-api)
- [Google Maps API](https://serpkite.com/apis/google-maps-api)
- [Google Maps Scraper API](https://serpkite.com/apis/google-maps-scraper)
- [Google Places API](https://serpkite.com/apis/google-places-api)
- [Google Place Details API](https://serpkite.com/apis/google-place-details-api)
- [Google Reviews API](https://serpkite.com/apis/google-reviews-api)
- [Google Shopping API](https://serpkite.com/apis/google-shopping-api)
- [Google Scholar API](https://serpkite.com/apis/google-scholar-api)
- [Google Patents API](https://serpkite.com/apis/google-patents-api)
- [Google Knowledge Graph API](https://serpkite.com/apis/google-knowledge-graph-api)
- [Google Autocomplete API](https://serpkite.com/apis/google-autocomplete-api)
- [Webpage to Markdown API](https://serpkite.com/apis/webpage-to-markdown-api)
- [Custom Search JSON API](https://serpkite.com/apis/google-custom-search-api)
- [Rank Tracker API](https://serpkite.com/apis/google-rank-tracker-api)

## AI APIs

- [AI Search API](https://serpkite.com/apis/ai-search-api)
- [AI Data Extraction API](https://serpkite.com/apis/ai-data-extraction-api)
- [Deep Research API](https://serpkite.com/apis/deep-research-api)

## AI assistant APIs (2–4 credits per answer)

- [ChatGPT Scraper API](https://serpkite.com/apis/chatgpt-scraper-api)
- [Gemini Scraper API](https://serpkite.com/apis/gemini-scraper-api)
- [Copilot Search API](https://serpkite.com/apis/copilot-search-api)
- [Google AI Mode API](https://serpkite.com/apis/google-ai-mode-api)
- [Google AI Overview API](https://serpkite.com/apis/google-ai-overview-api)
- [Brave AI Answers API](https://serpkite.com/apis/brave-ai-answers-api)

## Web data APIs

- [Website Crawler API](https://serpkite.com/apis/website-crawler-api)
- [Web Extract API](https://serpkite.com/apis/web-extract-api)
- [Website Map API](https://serpkite.com/apis/website-map-api)
- [SERP Monitoring API](https://serpkite.com/apis/serp-monitoring-api)

## Other search engines (engine=)

- [Brave Search API](https://serpkite.com/apis/brave-search-api)
- [Brave News API](https://serpkite.com/apis/brave-news-api)
- [Brave Image Search API](https://serpkite.com/apis/brave-image-search-api)
- [Brave Video Search API](https://serpkite.com/apis/brave-video-search-api)
- [Bing Search API](https://serpkite.com/apis/bing-search-api)
- [Bing News Search API](https://serpkite.com/apis/bing-news-search-api)
- [Bing Image Search API](https://serpkite.com/apis/bing-image-search-api)
- [Bing Video Search API](https://serpkite.com/apis/bing-video-search-api)
- [Yahoo Search API](https://serpkite.com/apis/yahoo-search-api)
- [Yahoo News API](https://serpkite.com/apis/yahoo-news-api)
- [DuckDuckGo Search API](https://serpkite.com/apis/duckduckgo-search-api)
- [DuckDuckGo News API](https://serpkite.com/apis/duckduckgo-news-api)
- [DuckDuckGo Image Search API](https://serpkite.com/apis/duckduckgo-image-search-api)
- [DuckDuckGo Video Search API](https://serpkite.com/apis/duckduckgo-video-search-api)
- [Yandex Search API](https://serpkite.com/apis/yandex-search-api)
- [Yandex Maps API](https://serpkite.com/apis/yandex-maps-api)
- [Yandex Places API](https://serpkite.com/apis/yandex-places-api)

## Features

- [Web search API for AI agents](https://serpkite.com/features/web-search-api-for-ai-agents)
- [Multi-engine search](https://serpkite.com/features/multi-engine-search)
- [Automatic search fallback](https://serpkite.com/features/search-fallback)
- [Consensus metasearch](https://serpkite.com/features/consensus-search)
- [LLM-ready search results](https://serpkite.com/features/llm-ready-search-results)
- [Search with page content](https://serpkite.com/features/search-with-page-content)
- [Domain and date filters](https://serpkite.com/features/domain-filters)
- [Bulk SERP API](https://serpkite.com/features/bulk-serp-api)
- [Cited AI answers](https://serpkite.com/features/cited-ai-answers)
- [Structured data extraction](https://serpkite.com/features/structured-data-extraction)
- [Deep research](https://serpkite.com/features/deep-research)
- [AI visibility tracking](https://serpkite.com/features/ai-visibility-tracking)

## Integrations

- [MCP (Claude, Cursor, ChatGPT)](https://serpkite.com/integrations/mcp)
- [LangChain](https://serpkite.com/integrations/langchain)
- [LangGraph](https://serpkite.com/integrations/langgraph)
- [CrewAI](https://serpkite.com/integrations/crewai)
- [LlamaIndex](https://serpkite.com/integrations/llamaindex)
- [Vercel AI SDK](https://serpkite.com/integrations/vercel-ai-sdk)
- [Next.js](https://serpkite.com/integrations/nextjs)
- [Mastra](https://serpkite.com/integrations/mastra)
- [OpenAI Agents SDK](https://serpkite.com/integrations/openai-agents-sdk)
- [Pydantic AI](https://serpkite.com/integrations/pydantic-ai)
- [n8n](https://serpkite.com/integrations/n8n)
- [Make](https://serpkite.com/integrations/make)
- [Zapier](https://serpkite.com/integrations/zapier)
- [Dify](https://serpkite.com/integrations/dify)
- [Activepieces](https://serpkite.com/integrations/activepieces)
- [Flowise](https://serpkite.com/integrations/flowise)
- [Python](https://serpkite.com/integrations/python)
- [Node.js / TypeScript](https://serpkite.com/integrations/node)
- [Go](https://serpkite.com/integrations/go)

## Migration guides

- [From Google Custom Search API](https://serpkite.com/migrate/google-custom-search-api)
- [From Bing Search API](https://serpkite.com/migrate/bing-search-api)
- [From SerpApi](https://serpkite.com/migrate/serpapi)
- [From Tavily](https://serpkite.com/migrate/tavily)
- [From Brave Search API](https://serpkite.com/migrate/brave-search-api)

## Tools

- [SERP checker](https://serpkite.com/tools/serp-checker): See the live Google results for any query and country.
- [Rank checker](https://serpkite.com/tools/rank-checker): Find where a domain ranks in the top 100 for a keyword.
- [SERP token counter](https://serpkite.com/tools/serp-token-counter): Compare the LLM token cost of JSON, compact and Markdown SERPs.
- [CSE migration helper](https://serpkite.com/tools/cse-migration): Paste a Custom Search API URL, get the SerpKite equivalent.

## Comparisons

- [SerpKite vs Serper](https://serpkite.com/compare/serpkite-vs-serper)
- [SerpKite vs SerpApi](https://serpkite.com/compare/serpkite-vs-serpapi)
- [SerpKite vs Tavily](https://serpkite.com/compare/serpkite-vs-tavily)
- [SerpKite vs DataForSEO](https://serpkite.com/compare/serpkite-vs-dataforseo)
- [SerpKite vs Scrapingdog](https://serpkite.com/compare/serpkite-vs-scrapingdog)
- [SerpKite vs SearchApi](https://serpkite.com/compare/serpkite-vs-searchapi)
- [SerpKite vs Brave Search API](https://serpkite.com/compare/serpkite-vs-brave-search-api)
- [SerpKite vs Exa](https://serpkite.com/compare/serpkite-vs-exa)
- [SerpKite vs Firecrawl](https://serpkite.com/compare/serpkite-vs-firecrawl)
- [SerpKite vs Perplexity Search API](https://serpkite.com/compare/serpkite-vs-perplexity-search-api)
- [Serper alternatives](https://serpkite.com/compare/serper-alternatives)
- [SerpApi alternatives](https://serpkite.com/compare/serpapi-alternatives)
- [Best SERP APIs in 2026](https://serpkite.com/compare/best-serp-api)
- [Free SERP APIs](https://serpkite.com/compare/free-serp-api)

## Optional

- [Pricing](https://serpkite.com/pricing)
- [Playground](https://serpkite.com/playground)
- [Blog](https://serpkite.com/blog)
- [How we collect data](https://serpkite.com/legal/how-we-collect)
- [Terms of Service](https://serpkite.com/legal/terms)
- [Refund Policy](https://serpkite.com/legal/refunds)
- [Privacy Policy](https://serpkite.com/legal/privacy)